Low dynamic muscle strength and its associations with fatigue, functional performance, and quality of life in premenopausal patients with systemic lupus erythematosus and low disease activity: a case–control study

The study was conducted between January 2009 and January 2011. A single examiner analysed the medical records and conducted structured interviews with 240 patients who were being followed in the outpatient Clinic of Rheumatology at the Brasilia University Hospital (HUB, Brasilia/Brazil). The interviews included socioeconomic status (e.g., education, employment, and income). Disease activity was assessed by the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) [16]. The control group was recruited through email, leaflets, and posters. Healthy women were primarily matched by physical activity levels, age and physical characteristics (i.e., body weight and body fat). The participants provided signed informed consent. The study was approved by the local ethics committee and was in accordance with the Helsinki Declaration of 1975, as revised in 1983.

All participants were premenopausal women. The SLE patients met the American College of Rheumatology (ACR) criteria [16], with stable disease (i.e., no flare-ups or changes in medication for at least 3 months before entering the study) [17]. Additionally, all participants had not engaged in regular exercise programs for at least six months prior to the study [10].

The following exclusion criteria were applied: SLEDAI > 5 (n = 19), serum creatinine ≥ 265 mmol/l, myositis, haematocrit ≤ 30%, nephritis and/or leukopenia (n = 13), history of myocardial infarction, cardiomyopathy, hypertension and/or the use of beta-blockers (n = 18), type 2 diabetes mellitus (n = 6), neurological diseases (n = 4), hypothyroidism (n = 5), fibromyalgia (n = 23), osteoporosis (n = 6), rheumatoid arthritis (n = 3), Sjögren’s syndrome (n = 2), cancer (n = 1), age < 18 years (n = 2) and > 45 years (n = 39), residence located far (other state) from the research centre (n = 40), body mass index (BMI) < 18 kg/m² (n = 1) and > 30 kg/m² (n = 8), smoking (n = 10), pregnancy (n = 1), and engaged in regular exercise (n =14).

The participants who met the inclusion criteria visited the laboratory on three different occasions in 48 to 72-hour intervals at the same time (2–4 pm). Two days prior to the tests, the participants were recommended to avoid intensive exercise, caffeine or alcohol intake. The evaluations were not performed during the menstrual period. On the first day, fatigue symptoms, quality of life, and physical activity level were assessed [15, 18, 19]. Additionally, anthropometric measurements and functional performance tests were performed. The patients were also familiarised with the one-repetition maximum strength tests (1-RM). On the second and third days, the 1-RM test and re-test were performed, respectively.

A single examiner evaluated height, weight, and BMI. Body fat was estimated using the skinfold measurement, as previously described [20].

All questionnaires were administered prior to the physical tests. The following fatigue scales were used: Fatigue Severity Scale (FSS) [1], which consists of a 9-question questionnaire with a score ranging from 1 to 7. Quality of life was assessed using the Short-Form Health Survey 36 (SF-36) [19], which consists of 36 items grouped into eight sub-domains (i.e., physical functioning, role-physical functioning, bodily pain, general health, vitality, social functioning, role-emotional functioning, and mental health). The SF-36 score ranges from 0 to 100, with a higher score indicating better health-related quality of life. The level of physical activities of daily living was evaluated using the short version of the International Physical Activity Questionnaire (s-IPAQ) [18], which consists of questions regarding the frequency (i.e., days per week) and duration (i.e., minutes per day) of occupational and recreational activities of daily living and structured exercise programs according to the physical activity level. The subjects were classified into three categories: active, irregularly active, and inactive.

Physical function was assessed through the following battery of tests. The 30-s chair stand test evaluated the number of times that a subject was able to stand from a standard chair and sit down again in 30 seconds [21]. The 30-s arm curl test assessed upper-body muscle function by the number of arm curl repetitions performed with 2-kg dumbbell for 30 seconds [21]. The handgrip strength test (Takei Kiki Kogyo Co., Ltd., Japan) evaluated the maximal isometric strength of the dominant hand using a calibrated dynamometer. The volunteers stand erect holding the dynamometer parallel to the side, with the dial facing away from the body. Each participant performed the test twice, with a 1-minute rest period between the measurements. The best value was chosen for the analysis. The grip position of the TKK dynamometer was adjusted to the individual’s hand size [22]. The Timed Up and Go (TUG) test assessed the time that a subject required to rise from a standard arm chair, walk 3 meters away, turn, return, and sit down again [23]. The one foot balance test with eyes closed assessed balance by having subject stand on one foot with eyes closed for up to 30 seconds [24]. The sit and reach test evaluated flexibility using the modified chair sit-and-reach test, as previously described [25].

The 1-RM test was used to determine the dynamic muscle strength of the upper- and lower-limbs [26] using conventional weight machines (Johnson Health Technologies, Taiwan) [5, 27‐29]. Prior to the 1-RM test, two light warm-up sets were performed in two-minute intervals. Then, the participants were given up to five attempts to achieve the 1-RM load (i.e., the maximum weight that could be lifted once using proper technique), with a five-minute interval between the attempts. The 1-RM tests were conducted for leg press, chest press, leg extension, lat pulldown, and leg curl exercises. The strength tests were performed on two different days and were separated by a 48–72 minute period, which allowed the calculation of the test-retest reliability (i.e., intra-class coefficient [ICC]) for both groups [28].

Sample size calculation was cited in a previous study [30] and assumed an effect size of 0.97 between groups; thus, a minimum sample of 25 volunteers for each group was required to provide 90% power (5% significance). The non-paired Student's T-test or Mann–Whitney U-test was used to compare the groups. Physical activity levels and socioeconomic status data were analysed by Pearson's chi-squared test.

The forward stepwise linear regression model was used to investigate the relationship between muscle strength (as the dependent variable) versus functional performance (i.e., 30-s chair stand test and handgrip tests), fatigue (i.e., FSS score), and quality of life (physical functioning subscale from SF-36) for the SLE patients. The regression model was applied as the 1-RM-total (muscle strength: sum of total load lifted in the 1-RM tests [i.e., leg press + leg extension + leg curl + chest press + lat pulldown]), according to Ruiz et al. [5]. Normally distributed data are expressed as the mean±SD, and non-normally distributed data are expressed as median and interquartile range. The significance level was set at 5%. The analyses were performed using the software SAS for Windows 9.2 (SAS Institute Inc., Cary, NC, USA).

Twenty-five premenopausal SLE patients with a low disease activity (SLEDAI = 1.5 ± 1.2, range = 0–5, 9 of 25 patients) and a disease duration of 5.3±4.6 years (range = 1–20 years) participated in this study. The patients were taking corticosteroids (21/25 [84%], dose = 6.07 ±2.1 mg/day, range = 5–20 mg/day), azathioprine (8 of 25 patients [32%], dose = 87.50 ±46.8 mg/day, range 50–200 mg/day), chloroquine diphosphate (17/25 [68%], dose = 205.88 ± 66.4 mg/day), and hydroxychloroquine (2 of 25 patients [8%], dose = 400 ±0.0 mg/day). The primary characteristics of the SLE patients and the matched healthy controls are shown in Table 1.

The SLE patients had lower educational level than the controls, P < 0.05). The groups did not differ regarding other socioeconomic status variables (P > 0.05) (Table 2).

When compared with the controls, the SLE patients had significantly higher FSS score (P < 0.01). The SLE patients also had poorer quality of life parameters when compared with the controls, (all P < 0.05), except for vitality and bodily pain domains (both P > 0.05) (Table 3).

When compared with the controls, the SLE patients had a significantly lower functional performance in general (handgrip test = -10.35%, TUG test = - 5.94%, 30-s chair timed-stand test = - 18.60%, 30-s arm curl test = - 16.58%, all P < 0.05), except for balance and flexibility (both P > 0.05) (Table 3).

The ICC for the 1-RM test was 0.98 (CI 0.60, 0.99) and 0.99 (CI 0.86, 0.98) for the SLE and control groups, respectively. The SLE patients had a significantly lower 1-RM than the controls in all exercises (1-RM = -25.63%, leg extension = -11.19%, leg curl = -15.71%, chest press = -18.33%, lat pulldown = -13.56%, 1-RM [total load] = - 18.12%, 1-RM/relative [total load/body weight = - 17%, all P < 0.05]) (Table 4).

The final model for dynamic muscle strength, which included the functional performance tests (handgrip and timed chair stands), SF-36 (physical role functioning and emotional role functioning scores), and FSS scores, accounted for 52% of the variance in dynamic muscle strength (F= 5.62, P = 0.02, VIF < 10, Table 5). Unexpectedly, physical role functioning was inversely related to 1-RM in the final model. However, physical role functioning seemed to be a weak independent factor in 1-RM change (R² = 0.14; = -0.38), therefore the role of this variable as a predictor in the final model must be interpreted with caution. The errors in the model were independently distributed (Durbin-Watson = 2.19), and multicollinearity was not detected (VIF = 1.0).

No adverse events were recorded during the experimental period. Additionally, none of the patients reported joint pain at time of testing.